Skin-Interfaced and Textile-Integrated Wearable Systems for Embodied Interaction and Digital Health

Abstract

Rapid advancements in Artificial Intelligence (AI), the Internet of Things, and 5th generation (5G) communication are ushering in a new era of integration between information technology and human society, profoundly transforming human-machine interaction and healthcare. In human-machine interaction, embodied interactions, exemplified by robotics, Virtual Reality (VR), and Augmented Reality (AR), are shifting away from traditional control-and-feedback paradigms toward seamless, natural, and immersive user experiences. In healthcare, digital health, which is characterized by continuous, proactive, and personalized care, is shifting traditional medicine from reactive treatment to preventive and precision interventions. Wearable systems, which serve as a physical medium connecting humans to the digital world and as a key node linking individuals to digital health, have become a central theme in advancing next-generation information technology and biomedical engineering. This thesis reports on developing basic components for wearable systems, including sensors and actuators, and further details their integration into skin-interfaced and textile-integrated systems. More critically, the potential applications of these wearable systems in embodied interaction and digital health were demonstrated. The main contributions of this thesis are summarized as follows:

First, various high-performance sensors were developed for human monitoring, including kinematic and electrophysiological sensing. Specifically, a durable e-textile strain sensor was developed based on a nanocomposite yarn strain sensor, fabricated by coating graphene and silver nanowires (AgNWs) onto a thin polyurethane (PU) yarn. The yarn strain sensor was then thermo-laminated into thermoplastic polyurethane (TPU) sheets, yielding an e-textile strain sensor that can be unobtrusively attached to or sewn into clothing. Moreover, a skin-interfaced liquid metal (LM)-based strain sensor was developed, functioning as an electronic skin for full finger joint monitoring and gesture classification. The LM-based strain sensor was fabricated through a magnet-induced pattern method, achieving a gauge factor (GF) of 5.17 at 300% strain. Furthermore, a thin (300 μm), low sheet resistance (0.481 Ω·sq^-1), and mechanically robust on-skin electrode was fabricated by drop-casting an AgNWs network onto an elastic polydimethylsiloxane (PDMS) substrate for detecting surface electromyography (sEMG) signals. This on-skin electrode recorded sEMG signals from forearm muscles with a higher signal-to-noise ratio (SNR) than commercial Ag/AgCl electrodes.

Second, two types of electromagnetic actuators with high haptic stimulation density were developed for haptic feedback in embodied interaction. In detail, an electromagnetic actuator with a compact size (2.2 mm thickness, 10 mm diameter) was developed based on a copper coil. The thickness of the magnet-carrying soft membrane was optimized to 0.23 mm, tuning the actuator’s resonant frequency to 210 Hz, nearing peak human sensitivity frequency. Besides, user tests illustrated that the haptic stimulation was perceived at voltage inputs of 0.14-0.73 V on the fingertip and 0.10-0.42 V on the palm. Subsequently, an elastomeric haptic electromagnetic actuator (EHEA) based on a dense LM microcoil was developed. The dual-layer LM microcoil was fabricated using a vacuum-assisted infiltration process in laser-ablated microfluidic channels. A 22-turn microcoil was found to optimally balance magnetic flux density and Joule heating. Due to the high magnetic flux density generation efficiency of the LM microcoil (19.86 T·W^-1·m^-2), users could perceive haptic stimulation at power inputs of only 0.809-2.369 mW. The low perceptual power mitigates the high power consumption of soft electromagnetic actuators used for haptic feedback.

Third, various wearable systems were developed based on the high-performance sensors and actuators described above, and their utility for embodied interaction was validated. In detail, a haptic intelligent motion tracking (HIMT) glove was created by integrating the e-textile strain sensors and copper coil-based electromagnetic vibrators with inertial measurement units (IMUs). The glove aims to enhance virtual interactions by mapping upper-limb motions onto a virtual human for natural human-virtual interaction. The glove’s performance was validated through interactive physiotherapy and clinical diagnosis. In addition, a haptic finger sleeve interface was developed by integrating the EHEAs and LM-based strain sensors. This interface facilitates a closed-loop haptic control and feedback system, which enables the teleoperation of a robotic arm with stiffness perception. It also demonstrated the remote palpation of pathological tissues in sigmoid colon cancer by employing teleoperation of an endoscope with haptic feedback.

Fourth, an e-textile knee pad and a multi-modal sensor glove were developed for assessing upper- and lower-limb functions, based on e-textile strain sensors and other sensors. The systems aim to design textile-integrated wearable systems for functional assessment in older adults. The multi-modal sensor glove integrates an e-textile strain sensor, a force sensor array, and an IMU to comprehensively evaluate hand strength, finger dexterity, and hand stability. A significant correlation analysis was performed between various metrics from these tests and older adults’ appendicular skeletal muscle mass index (ASMI). Furthermore, older adults’ lower-limb movement performance was assessed using an e-textile knee pad during sit-to-stand (STS) motions. Users were grouped by age, body mass index (BMI), and gender. A significant correlation was found between the amplitude and stability of fast speed STS movements and the age, gender, and BMI of the older adults. These findings underscore the applicability of the textile-integrated wearable systems and highlight their potential for long-term health monitoring within the geriatric population.

In summary, this thesis systematically investigated wearable systems for embodied interaction and digital health, from the design and fabrication of fundamental components, sensors and actuators, to their integration into skin-interfaced and textile-integrated wearable systems. The integrated wearable systems enabled seamless human-machine interaction and effective multi-modal health monitoring. This thesis provides tangible solutions to key challenges in wearable technology, paving the way for next-generation wearable systems with advanced performance and multi-functionality for transformative applications in immersive embodied interaction and personalized digital health.

Date of Award	20 Jan 2026
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Wai Chiu King LAI (Supervisor)